Vol. 41 2015 No. 4 DOI 10.5277/epe150406
HANNA OBARSKA-PEMPKOWIAK1, MAGDALENA GAJEWSKA1, EWA WOJCIECHOWSKA1, KATARZYNA KOŁECKA1
SEWAGE GARDENS – CONSTRUCTED WETLANDS
FOR SINGLE FAMILY HOUSEHOLDS
During the last 30 years the increase of interest in low-cost decentralized and small wastewater treatment plants (WWTP) is observed in Europe. For this purpose, sand filters, stabilization ponds, mini reactors with activated sludge, trickling filters and treatment wetlands are used. Among the listed facilities, treatment wetlands (TWs) offer effective and reliable treatment efficiency, being at the same time low-cost and simple in operation. This is a sustainable treatment technology which does not gen-erate secondary sewage sludge. In the paper, design concept and performance of nine TWs constructed in 2009 for single-family households in Kashubian Lake District is discussed. The treatment efficiency of three different TW configurations is compared. The findings of the study provide a basis for future design and implementation of TWs in agricultural areas.
1. INTRODUCTION
Treatment wetlands (TWs) have already found a wide application in municipal man-agement all over the world. TWs are first of all used for second stage treatment of do-mestic wastewater but also for the treatment of industrial wastewater (e.g. from wineries and dairies), stormwater, landfill leachate as well as renaturalization of treated effluent from conventional WWTPs [1–3]. In many European countries, TWs are advised as the best available technology (BAT) for single family households and local treatment facil-ities for less than 2000 person equivalent (pe) [4]. In spite of the fact that the method has been used for over forty years, the optimal design and configuration of TWs to guarantee both reliability and pollutant removal is still a contentious issue. According to Langergraber [5], the effluents of a one-stage vertical subsurface flow (VSSF) bed of a unit surface area equal to 4 m2/pe and organic matter loading equal to 20 g/(m2·d), meet rigorous Austrian
_________________________
1Gdańsk University of Technology, Faculty of Civil and Environmental Engineering, ul. Narutowicza 11/12, 80-233 Gdańsk, Poland, corresponding author E. Wojciechowska, e-mail: esien@pg.gda.pl
outflow standards (below 90 mg COD/dm3 and 25 mg BOD
5/dm3), regardless of the
sea-son and air temperature. Simple guidelines for designing TWs for less than 30 pe with single VSSF bed were introduced in Denmark in 2005 [6]. In France, two sequential VSSF beds have been successfully used for the treatment of raw sewage (without pri-mary mechanical treatment) for over 20 years [7, 8]. According to Molle et al. [8] the unit area should be equal to 1.2 m2/pe for the first bed and only 1.0 m2/pe for the second
bed. Such configuration of VSSF beds allows the reduction of pollutant concentrations to the following levels: COD – 60 mg/dm3, TSS – 15 mg/dm3, Kjeldahl nitrogen
– 8.0 mg/dm3. Molle et al. [8] recommended that the hydraulic loading of the beds
work-ing in batches should be below 600 mm/d. In Norway, a very effective primary treat-ment of wastewater is applied. A three-chamber septic tank is recommended for TWs [9]. Effluent from the tank is first pre-treated by the biofilter (pre-filter) before entering the wetland bed. The pre-filter is a vertical flow bed filled with LECA material (light-weight expanded clay aggregates) which has very good filtration properties. Another function of the pre-filter is aeration of sewage. The last step of domestic wastewater treatment occurs in a HSSF bed (horizontal subsurface flow). This bed is mainly con-structed for removal of phosphorus and pathogens. For efficient phosphorus removal in TWs, the use of FiltraliteP as a filter media with high P-binding capacity is
recom-mended [9]. Effective phosphorus removal requires a minimum retention time of 7–10 days in a HSSF bed. Due to the cold climate, HSSF beds in Norway are deeper (0.9 m) than it is recommended in other European countries [10].
The European experience also shows that TWs can be successfully used for the stabilization and disposal of sewage sludge in the so called sludge treatment reed beds (STRBs) [11]. The STRB is a system consisting of beds made of concrete, prefabricates or soil screened with an impermeable membrane and drainage pipes for collecting the leachate at the bottom. The beds are filled with gravel (for effective sludge dewatering and bed ventilation) covered by sand (to support plant growth). Macrophytes (usually
Phragmites australis) are implemented to grow on overlying layers of deposited sludge
(with low content of dry matter, ca. 0.5–1%) [12–14]. Estimated costs of dewatering sludge in STRBs are relatively low and constitute only 5–10% of the overall operational costs of conventional methods of sludge utilization [15].
Wetland technology could also be used for solving problems in wastewater and sludge management in rural areas. It is called green technology, due to the use of vege-tation and its low energy demand (very often systems work without any energy con-sumption). TWs create a similar environment to natural wetlands, promoting biodiver-sity. For these reasons, this treatment method is regarded environmentally friendly and sustainable [16]. Moreover, TWs fit perfectly within a village landscape and can easily be arranged as sewage gardens without producing any negative impacts (like odors or mosquitoes), even in close proximity to households.
According to the EU directive, the problem of wastewater treatment in non-urban-ized areas must be solved by 2015. To meet these demands, three different concepts of
approaching wastewater treatment and sludge utilization in rural areas were developed by Polish researchers from the Gdansk University of Technology in cooperation with Norwegian researchers from the Bioforsk, Soil and Environment Division in Aas within the research project Innovative Solutions for Wastewater Management in Rural Areas. These concepts followed the main idea of the research project aimed at implementing innovative and sustainable technologies for management and treatment of domestic wastewater and sludge in rural areas.
The aim of this paper is to present the results of an innovative wastewater and sludge management system for single households that was designed and implemented in 2009 in the scope of the research project Innovative Solutions for Wastewater Management
in Rural Areas (financed by Polish Ministry of Science and Higher Education
E033/P01/2008/02 and EOG Financial Mechanism, and Norwegian Financial Mecha-nism PL0271). The proposal of an innovative sanitary system is based on the idea of a closed cycle of matter in the environment. The compounds of nutrient substances: N, P, K present in wastewater should be used as soil fertilizers. Wastewater from single households is treated at individual treatment plants. One of the benefits of TWs is no excessive (secondary) sludge generation during wastewater treatment. Primary treat-ment takes place in septic tanks. Then the wastewater is discharged to treattreat-ment wet-lands working in one of three configurations. According to the project, sewage sludge gathered in septic tanks was to be periodically removed and discharged to reed treatment beds, where intensified natural dewatering and stabilization would take place. The fil-trate generated in the dewatering process is recirculated to TWs and treated together with wastewater (Fig. 1). The dewatered and stabilized sewage sludge would become a valuable humus substance, that could be used as soil fertilizer at the farm lands. In this way, the matter cycle is closed. The sewage treatment technology as well as sludge processing is simple in operation.
Fig. 1. The concept of innovative wastewater and sludge management for a single household
The innovation of the proposed solution was the discharge of the effluent to a pol-ishing pond to increase the retention of water in a small catchment area as well as the
application of sludge treatment reed beds (STRBs) for primary sludge produced in the septic tank. In practice, it turned out that farmers had a negative attitude towards the use of STRBs. The facilities were built according to the design project, however none of the farmers involved in the project ever decided to use them. Instead, the sludge from septic tanks is periodically pumped out and transported to a central wastewater treatment plant.
2. METHODS
Study facilities. Three different configurations of SSF systems were applied to test
whether vertical or horizontal flow beds perform better in Polish countryside conditions. In all configurations, a septic tank was applied as a mechanical treatment and the last stage of treatment was performed in a polishing pond. In configuration I, a single VSSF bed with a unit area of 4m2/pe was applied while in configuration II sequentially
work-ing VSSF beds were applied with the total unit surface area of 2.5 m2/pe. The
configu-rations with a single VSSF bed and two VSSF beds working in a series were tested to investigate how double contact time affects the efficiency of pollutant removal. Config-uration III was designed according to the Norwegian scheme and consisted of a pre-filter followed by a HSSF bed. Instead of LECA, a local material – Pollytag made of fly ashes from thermal-electric power plants was used (Table 1).
Table 1. Characteristics of the analyzed single family TWs
Configuration
HRT in primary tank [day]
Unit area/pe
[m2] Filling material of the bed
Pond surface area [m2]a I with single VSSF maximum 3 3.0 20 cm – drainage layer 16–32 mm 40 cm –filtration layer 2–8 mm up to 10 cm – mixture of peat and wooden bark
Minimal 1 m2/pe with the depth maximum 1.2 m II with two VSSF 5–7 days I VSSF – 1.5 II VSSF – 1.0 20 cm – drainage layer 16–32 mm 20 cm – filtration layer 8–16 mm 20 cm – filtration layer 2–8 mm up to 10 cm – mixture of peat and wooden barkb
III with pre-filter and HSSF
3–7 days 5.0
distribution and collection zone 16–32 mm
main filtration layer 2–8 mm
aMinimal designed (recommended) dimension – all farmers decided to build ponds as recipients of treated sewage.
bThe same layers for I and the II stage beds.
An important feature of the implemented wastewater treatment plants is that all of them have been implemented in 2009 as full-scale facilities under real field-operational
conditions. Therefore the outcomes on the removal of wastewater pollutants truly reflect the treatment efficiency of these facilities.
Sampling and laboratory analyses. The sampling period was divided into two
sub-periods: I – in years 2010 and 2011 with seven sampling events and II – from October 2012 to May 2013 with three sampling events. The second period should be regarded a post-vegetation one. The averaged samples of wastewater inflowing to the HSSF/VSSF beds (after septic tanks) were collected from the collecting wells (Fig. 1). The size of the wells ensured 6 h of wastewater retention. Outflow samples were collected from the ponds. In the collected samples COD, TSS, total nitrogen (TN) and total phosphorus (TP) were determined. The analytical procedures followed the recommendations of the Hach Chemical Company and Dr Lange GmbH. The analyses were performed accord-ing to Polish Norms and guidelines as set out in the Polish Environmental Ministry Regulation of 24 July 2006 [17].
3. RESULTS
3.1. POLLUTANTS CONCENTRATION IN WASTEWATER IN SINGLE FAMILY TWs
Figures 2 and 3 present the average concentrations of pollutants in wastewater at the inflow and at the outflow of the analyzed TWs in both monitoring periods (I and II). The results are given for three TW configurations. In each configuration three facilities (described with Nos. 1, 2 and 3) are shown with different shaded bars.
The quality of wastewater delivered to the TWs after mechanical treatment in the septic tanks differed significantly in regards to COD, TSS, TN and TP concentrations (Fig. 2). In both sampling periods, the highest concentrations of COD were observed in the facilities operating according to configuration III. The concentrations of COD in configuration III varied from 900 to 1300 mg O2/dm3 while in configurations I and II
they were below 800 mg O2/dm3. The high concentrations of COD were not associated
with particulate organic matter since the concentrations of TSS were much lower and varied from 160 to 450 mg/dm3. Thus most of organic matter incoming into TWs was
in the soluble or colloidal form. It was also observed that in the second sampling period the concentration of TSS decreased significantly, most probably due to better mainte-nance of septic tanks, while the COD concentrations remained high.
Wastewater delivered to the treatment wetlands was characterized by a high concentra-tion of total nitrogen, from 60 to 160 mg TN/dm3, in both periods (Fig. 2c). These
concen-trations were two–three times higher in comparison to the values reported by Vymazal [18] and Heistad et al. [19] (36.3 to 77.5 mg TN/dm3). Similar high concentrations of nitrogen
were observed in septic tank effluents in the Podlasie region of Poland [20, 21]. Such dif-ferences in TN influent concentration indicate that in some cases domestic wastewater was contaminated with liquid manure or other liquid wastes from the farmyards.
Fig. 2. Concentrations of pollutants discharged to single family TWs a) TSS, b) COD, c) TN, d) TP; 1–3 – numbers of facilities in each configuration
0 100 200 300 400 500
Config I Config II Config III Config I Config II Config III
I period II period TSS inflowing, mg/dm3 1 2 3 a) 0 400 800 1200
Config I Config II Config III Config I Config II Config III
I period II period COD inflowing, mg/dm3 1 2 3 b) 0 50 100 150 200
Config I Config II Config III Config I Config II Config III
I period II period TN inflowing, mg/dm3 1 2 3 c) 0 5 10 15 20
Config I Config II Config III Config I Config II Config III
I period II period
TP inflowing, mg/dm3
1 2 3
Fig. 3. Concentrations of the pollutant in the effluent from single family TWs: a) TSS, b) COD, c) TN, d) TP. 1–3 – numbers of facilities in each configuration
0 50 100 150
Config I Config II Config III Config I Config II Config III
I period II period TSS out‐flowing, mg/dm3 1 2 3 a) 0 50 100 150 200 250 300
Config I Config II Config III Config I Config II Config III
I period II period COD out‐flowing, mg/dm3 1 2 3 b) 0 20 40 60 80
Config I Config II Config III Config I Config II Config III
I period II period TN out‐flowing, mg/dm3 1 2 3 0 3 6 9 12 15
Config I Config II Config III Config I Config II Config III
I period II period
TP out‐flowing, mg/dm3
The highest variability in inflow total phosphorus concentrations was observed at facilities working in configuration I: from 5 to 18 mg TP/dm3 in the first period and for
configuration II during the second period. For other configurations, the quality of deliv-ered wastewater was more constant. The range of TP concentrations was 14.0–17.5 mg TP/dm3 (Fig. 2d).
The quality of treated wastewater is shown in Fig. 3. Although the concentrations of pollutants decreased, the sewage samples collected from the last stage of treatment (the pond) in many cases did not fulfil the requirements of the Regulation of Environ-mental Ministry [17] during the first monitoring period. The relatively low quality of the treated effluent in this period was likely be the result of a short period of the TWs operation (poor development of roots and rhizomes as well as biofilms) and of the short period of wastewater retention in the pond [20]. During the second period (post-vegeta-tion one) the quality of the effluent improved significantly and in many cases met the above mentioned requirements. Effluents from single family TWs working in configu-rations I and II met all the requirements except for concentconfigu-rations of total phosphorus. Effluent quality in configuration III was within the required value (50 mg/dm3) for
TSS concentrations only. None of analyzed TWs ensured TP effluent concentration be-low the required 5 mg/dm3.
3.2. REMOVAL EFFICIENCY
Within most single family TWs, the removal of pollutants was quite effective, de-spite high their concentrations at the inflow. The efficiencies of analyzed pollutant re-moval in both monitoring periods are presented in Tables 2 and 3.
T a b l e 2 Total removal efficiency [%] of suspended solids and organic matter in single family TWs
in both monitoring periods
Facility TSS COD Configuration I II III I II III First period 1 89.1 85.5 43.6 84.1 68.6 79.9 2 58.3 72.3 79.9 60.0 83.1 82.5 3 67.1 78.0 45.8 53.1 78.6 83.3 Mean 71.5 78.6 56.4 65.7 76.8 81.9 Second period 1 93,5 79.9 83.7 91.5 81.5 85.5 2 90.7 86.8 81.2 91.2 84.7 85.8 3 82.0 91.3 86.9 86.3 76.5 85.9 Mean 88.8 86.0 83.9 89.6 80.9 85.7
T a b l e 3 Removal efficiency [%] of nutrient compounds (TN and TP)
in single family TWs in both monitoring periods
Facility TN TP Configuration I II III I II III First period 1 70.6 75.9 49.7 68.2 67.8 57.6 2 55.9 79.0 69.9 65.4 61.4 54.7 3 55.6 54.3 55.8 29.1 64.0 50.0 Mean 60.7 69.7 58.5 54.3 64.4 54.1 Second period 1 82.3 73.8 63.6 21.2 1.9 30.1 2 78.5 73.5 64.7 39.7 21.6 26.2 3 75.0 77.8 68.8 5.6 13.7 37.4 Mean 78.6 75.0 65.7 22.1 12.4 31.2
Generally the efficiency of pollutant removal (except phosphorus) improved with time in all configurations. The highest improvement was observed for configuration I with a single VSSF bed with the lowest loading rate. In the first period, TSS were gen-erally more efficiently removed in facilities with vertical flow beds while in the second one, the removal efficiency of TSS in all facilities of configuration III improved signif-icantly and was similar to those in other configurations. In the first monitoring period, configuration III was the most effective in respect to COD removal while in the second monitoring period COD was removed with the highest efficiency in configuration I.
Although the efficiency of TN removal in configuration III increased from 58.5% in the first monitoring period to 65.7% in the second one, it still remained lower than in the two other configurations. TP removal was very effective in the first monitoring pe-riod due to high sorption capacity of the substrate used. When the sorption capacity decreased in the second period, the removal occurred only due to bioaccumulation and did not exceed 30% of the incoming phosphorus load [2].
4. DISCUSSION
The pollutant concentrations in the wastewater discharged to the single family TWs were much higher than reported by numerous authors [18, 19, 22, 23]. Such high inflow concentrations seem to be typical of Polish rural areas and could be caused either by improper maintenance and operation of septic tanks or the inflow of high strength wastewater (manure, run-off from the fields or leakages from farmyard) and small water
consumption, at least in some cases [21, 24]. The observations based on the results of this study are in accordance with monitoring results in Europe. Many authors report that VSSF beds are more effective in terms of pollutant removal and provide effluents of better quality, as it was also confirmed in this study [23, 25–27].
The obtained results confirm that TWs were not successful in removing TP from wastewater due to a high concentration of P-compounds in volatile suspended solids. TWs are adapted to the removal of inorganic phosphorus compounds coming from de-tergents, while organic phosphorus should be retained together with the suspended sol-ids in the mechanical treatment unit.
A comparison of all three configurations indicates superior performance of config-urations with VSSF beds. Configuration I consisted of a single VSSF bed with a larger unit surface area in comparison to two sequential VSSF beds working in configuration II. In both configurations the beds were followed by a pond. The final pollutant concen-trations within the effluent from both configurations were similar and fulfilled the re-quirements of the Regulation of Environmental Ministry [17]. Since the contact time in sequentially working VSSF beds was twice longer than in a single VSSF bed with big-ger unit surface area, it can be concluded that an elongation of the contact time has a greater positive impact on treatment results than the application of a larger surface area single unit. This finding is in accordance with the newest tendency to minimize the unit area of TWs but applying two or even more treatment stages in TWs [2, 21, 24, 27].
5. CONCLUSIONS
The application of TWs for single-family effluent is an effective and sustainable solution for wastewater treatment in rural areas.
The TWs operated in Poland receive much higher concentrations of pollutants in comparison to the TWs operated in Europe and the USA.
In first two years of operation, satisfactory treatment effectiveness of BOD 64.0–92.0%, TN 44.0–77.0%, TP 24.0–66.0% was observed.
The efficiency removal of organics and TN, except for TP, increased between 12–20% after three years.
After three years of operation, the best quality of effluent was obtained in single VSSF’s with a larger unit surface area and thus smaller loading ahead of sequentially working VSSF beds.
A negative attitude of the farmers can be an obstacle in the implementation of sludge treatment reed beds (STRBs). Thus STRBs at central WWTPs, receiving sludge from individual single-family treatment plants, should be considered instead.
ACKNOWLEDGMENTS
Financial support is acknowledged from S61-Strategies for sustainable communal waste-water man-agement in the Baltic Sea Region (SUWMAB) of the Seed Money Facility of the European Union Strategy for the Baltic Sea Region (EUSBSR Seed Money Facility).
REFERENCES
[1] MASI F.,CAFFAZ S.,GHRABI A., Multi-stage constructed wetlands systems for municipal wastewater
treatment, Water Sci. Technol., 2013, 67 (7), 1590.
[2] VYMAZAL J., Constructed wetlands for wastewater treatment, Water, 2010, 2, 530.
[3] MEYER M.,MOLLE P.,ESSER D.,TROESCH S.,MASI F.,DITTMER U., Constructed wetlands for
com-bined sewer overflow treatment. Comparison of German, French and Italian approaches, Water,
2013, 5, 1.
[4] GRIFFIN P.,PAMPLIN C., The advantages of a constructed reed bed based strategy for small sewage
treatment works, Water Sci. Technol., 1998, 38 (3), 143.
[5] LANGERGRABER G., Simulation of the treatment performance of outdoor subsurface flow constructed
wetlands in temperate climates, Sci. Total Environ., 2007, 380, 210.
[6] BRIX H.,ARIAS C.A., The use of vertical flow constructed wetlands for on-site treatment of domestic
wastewater. New Danish guidelines, Ecol. Eng., 2005, 25, 491.
[7] BOUTIN C.,LIENARD A.,ESSER D., Development of a new generation of reed-beds filters in France.
First results, Water Sci. Technol., 1997, 35 (5), 315.
[8] MOLLE P.,LIENARD A.,BOUTIN C.,MERLIN G.,IWEMA A., How to treat raw sewage with constructed
wetlands?. An overview of the French systems. Proc. 9th International Conference on Wetland System
for Water Pollution Control, Avignon, France, 2004, 11.
[9] NORVAR, Principles for Construction of Wetland Filters. Våtmarksfiltre, NKF og NORVARs VA/Miljø-blad, No. 49, 2001 (in Norwegian).
[10] PARUCH A.M.,MÆHLUM T.,OBARSKA-PEMPKOWIAK H.,GAJEWSKA M.,WOJCIECHOWSKA E.,OSTOJSKI A.,
Rural domestic wastewater treatment in Norway and Poland. Experiences, cooperation and concepts on the improvement of constructed wetland technology, Water Sci. Technol., 2011, 63, 776.
[11] PERUZZI E.,MACCI C.,DONI S.,MASCIANDARO G.,PERUZZI P.,AIELLO M.,CECCANTI B., Phragmites
australis for sewage sludge stabilization, Desalination, 2009, 246, 110.
[12] NIELSEN S.,Sludge treatment reed bed facilities. Organic load and operation problems, Water Sci.
Technol., 2011, 63 (5), 941.
[13] KOLECKA K.,OBARSKA-PEMPKOWIAK H., Operation of reed systems used to stabilisation of sewage
sludge, Polish J. Environ. Studies, 2009, 6, 60.
[14] KOLECKA K.,OBARSKA-PEMPKOWIAK H., Speciation of heavy metals in stabilized sewage sludge
ac-cording to innovative technology with reed basins application in Denmark, Polish J. Environ. Studies,
2010, 2, 116.
[15] OBARSKA-PEMPKOWIAK H.,TUSZYNSKA A.,SOBOCINSKI Z.,Polish experience with sewage sludge
de-watering in reed systems, Water Sci. Technol., 2003, 48 (5), 111.
[16] PRICE T.,PROBERT D.,Role of constructed wetlands in environmentally-sustainable development,
Appl. Energy, 1997, 57 (2–3), 129.
[17] Environment Ministry Regulation according limits for discharged sewage and environmental protec-tion from 24 July 2006 (Dz.U. No. 137, item 984).
[18] VYMAZAL J., Horizontal sub-surface flow and hybrid constructed wetlands systems for wastewater
[19] HEISTAD A.,PARUCH A.M.,VRÅLE L.,ADAM K.,JENSSEN P.D., A high-performance compact filter
system treating domestic wastewater, Ecol. Eng., 2006,. 28 (4), 374.
[20] OBARSKA-PEMPKOWIAK H.,GAJEWSKA M.,WOJCIECHOWSKA E., Efficiency of wastewater treatment in
single-family constructed wetlands in Kaszuby Lake District, Ann. Set Environ. Prot., 2013, 15, 81.
[21] GAJEWSKA M.,KOPEĆ Ł.,OBARSKA-PEMPKOWIAK H., Operation of small wastewater treatment
facil-ities in a scattered settlement, Ann. Set Environ. Prot., 2011, 13 (1), 207.
[22] STEER D.,FRASER L.,BODDY J.,SEIFERT B., Efficiency of small constructed wetlands for subsurface
treatment of single-family domestic effluent, Ecol. Eng., 2002, 18, 429.
[23] BRIX H.,JOHANSEN N.H., Treatment of domestic sewage in a two-stage constructed wetland design
principles, [in:] Nutrient Cycling and Retention in Natural and Constructed Wetland. Leiden,
J. Vymazal (Ed.), Backhuys Publishers, Leiden 1999, 155.
[24] GAJEWSKA M.,OBARSKA-PEMPKOWIAK H., Efficiency of pollutant removal by five multistage
con-structed wetlands in a temperate climate, Environ. Prot. Eng., 2011, 37 (3), 27.
[25] GHRABI A.,BOUSSELMI L.,MASI F.,REGELSBERGER M., Constructed wetland as a low cost and
sus-tainable solution for wastewater treatment adapted to rural settlements. The Chorfech wastewater treatment pilot plant, Water Sci. Technol., 2011, 63 (12), 3006.
[26] LANGERGRABER G.,PRANDTSTETTEN C.,PRESSL A.,HABERL R.,ROHRHOFER R., Removal efficiency of
subsurface vertical flow constructed wetlands for different organic loads, Water Sci. Technol., 2007,
56 (3), 75.
[27] MOLLE P.,PROST-BOUCLE S.,LIENARD A., Potential for total nitrogen removal by combining vertical flow
